The human glioma cell line SWO-38 was established by the Department of Pathology, Medical College of Jinan University (Guangzhou, China) (6) and grown in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% newborn bovine serum (TBD Biotechnology Development, Tianjin, China), 100 U/ml penicillin and 100 mg/ml streptomycin. The cells were cultured at 37°C in a humidified incubator with 5% CO₂ and were regularly examined using an inverted microscope.

The morphology of CT-treated and untreated cells was investigated during a time course of 48 h using an inverted microscope (TE2000; Nikon Corporation, Tokyo, Japan).

SWO-38 cells in culture flasks were treated with 20 ng/ml CT alone for 72 h or pretreated with the MAPK inhibitor PD0325901, the PI3K inhibitor LY294002 or DMSO (as a solvent control) for 2 h. The cells were harvested and washed with PBS. Cells (1×10⁶) were fixed in 70% ethanol at 4°C for 24 h, washed in PBS, resuspended in 0.5–1.0 ml PBS containing 50 mg/ml RNase A at 37°C for 15 min and incubated in 50 mg/ml propidium iodide (PI; Sigma-Aldrich, St. Louis, MO, USA) in the dark at room temperature for 15 min. Flow cytometry was performed using a FACScan flow cytometer (Becton-Dickinson, San Jose, CA, USA).

Cell migration and invasion capacity was examined using a 24-well Transwell plate with 8-mm pore polycarbonate membrane inserts, according to the manufacturer's instructions (Corning, Inc., Corning, NY, USA). Matrigel (14.8 μg/ml; BD Biosciences, Franklin Lakes, NJ, USA) was employed for the invasion assays and applied to the upper surface of the membranes. Cells (2×10⁵ and 1×10⁵/well for invasion and migration, respectively) were seeded in the top chamber in serum-free media and this was replaced with complete growth media for 20 h for invasion and 12 h for migration. The cells that invaded or migrated through the surface of the membrane were fixed with methanol and stained with 0.1% hexamethylpararosaniline. Migrating or invasive cells from three random microscopic fields for each filter were selected for cell counting.

The human PTEN ShortCut siRNA and negative siRNA (control) were purchased from Invitrogen. The coding strand sequence of the PTEN siRNA was sense, 5′-GGCGCUAUGUGUAUUAUUAdTdT-3′ and antisense, 3′-dTdTCCGCGAUACACAUAAUAAU-5′. SWO-38 cells with 30–50% confluence were transfected using Lipofectamine™ 2000 reagent (Invitrogen), according to the manufacturer's instructions. The inhibition of PTEN protein expression was assessed using western blot analysis 6 h after transfection. The cells were then incubated with CT for an additional 72 h for further analysis.

Protein samples were prepared with lysis buffer containing 50 mM Tris (pH 8.0), 150 mM NaCl, 1% NP-40, 0.1% SDS, 100 ng/ml phenylmethylsulfonyl fluoride (PMSF) and 1 mg/ml aprotinin. For western blot analysis, equal amounts of proteins were loaded onto 10% SDS-PAGE gels for electrophoresis and then transferred to nitrocellulose membranes (Pall Corporation, Cortland, NY, USA). The nitrocellulose membranes were immediately blocked with 5% non-fat milk in Tris-buffered saline (TBS)-T (0.05% Tween-20 in TBS) buffer for 1 h at room temperature and then incubated overnight at 4°C with primary antibodies against glial fibrillary acidic protein (GFAP; 1:100; LabVision, Fremont, CA, USA), PTEN (1:1,000, Cell Signaling Technology, Inc., Beverly, MA, USA), AKT and p-AKTser473 (1:1,000; Cell Signaling Technology, Inc.), respectively. Following incubation with horseradish peroxidase-labeled secondary antibody at room temperature for 1 h, the membranes were detected using SuperSignal West Pico Chemiluminescent Substrate (Pierce, Rockford, IL, USA). GAPDH (1:4,000; Millipore, Billerica, MA, USA) or β-actin (1:500; Proteintech Group, Chicago, IL, USA) were used to normalize the quantity of protein on the blots.

Data are presented as the mean ± standard deviation (SD) from three separate experiments. Statistical analysis was carried out by an independent samples t-test and one-way ANOVA using the statistical package SPSS 16.0. P<0.05 was considered to indicate a statistically significant difference.

The differentiation of human SWO-38 glioma cells towards an astrocytic type was characterized by a marked morphological transformation from a flat polygonal appearance to multiple elongated cytoplasmic processes, similar to those of mature astrocytes (Fig. 1A and B). Microscopic observation of SWO-38 glioma cells treated with 10 ng/ml CT revealed major alterations in their morphology. We further examined the expression of GFAP, a well-documented marker of mature astrocytes. As shown in Fig. 1C and D, western blot analysis demonstrated a significant upregulation of GFAP protein expression in CT-treated cells compared with the control cells. However, this upregulation was not dose-dependent or time-independent.

CT induced significant cell cycle arrest in SWO-38 cells. Incubation with 20 ng/ml CT for 72 h caused an accumulation of cells in the G₀/G₁ phase; the percentage of cells in this phase reached 73.01±2.55% compared with 65.24±3.74% in the control group (Fig. 2A and B). Furthermore, there was a significant decrease in the percentage of cells in the S phase. The percentage of the cells in the S phase was 17.19±2.53% in the CT-treated group and 25.58±3.18% in the control group. Western blot analysis confirmed that treatment with 20 ng/ml CT for 72 h downregulated the cyclin D1 protein levels (Fig. 2C and D). Based on these results, CT appears to block cell-cycle progression from the G₁ to the S phase.

CT-treated SWO-38 cells exhibited a significantly decreased invasion (Fig. 3A and B) and migration (Fig. 3C and D) capacity compared with the control cells (P<0.05; Fig. 3E).

To elucidate the underlying molecular mechanism of CT-induced SWO-38 glioma cell differentiation, we determined the expression of the main proteins involved in the PI3K/AKT signaling pathway. As shown in Fig. 4, p-AKTser473 activity decreased after SWO-38 cells were treated with 10 ng/ml CT for 72 h, whereas the AKT activity was not altered. Additionally, CT treatment upregulated GFAP and downregulated cyclin D1 expression in the SWO-38 cells.

To further investigate whether PTEN is required during the process of CT-induced SWO-38 cell differentiation, siRNA was used to selectively knockdown the PTEN gene. After 6 h of transfection, the PTEN level was markedly reduced compared with the control cells (Fig. 5F). After the addition of CT, the cells were incubated for a further 72 h. The differentiation effects of CT on cell morphology (Fig. 5A and B), GFAP expression (Fig. 5F), cell cycle distribution (Fig. 5C and D) and invasion and migration capacity (Fig. 5E) in SWO-38 cells were attenuated.

To provide additional evidence for the importance of the PTEN/PI3K/AKT pathway in SWO-38 cell differentiation, we assessed the effects of the MAPK inhibitor PD0325901 and PI3K inhibitor LY294002 on CT-induced SWO-38 cell differentiation. SWO-38 cells were incubated with LY294002 (100 μmol/l) or PD0325901 (40 μmol/l) for 2 h. CT was subsequently added and the cells were incubated for an additional 72 h. As shown in Fig. 6A and B, CT-induced morphological transformation was attenuated via the inhibition of PI3K activity by LY294002, while exposure to the MAPK inhibitor PD0325901 and CT did not alter PI3K activity (Fig. 6C and D). Furthermore, the upregulation of GFAP and p-AKTser473 and the downregulation of cyclin D1 induced by CT were reverted by LY294002, but not by PD0325901 (Fig. 6G and H). The cell cycle alterations (G₀/G₁ cell cycle arrest and accelerated S phase entry; Fig. 6E) and cell migration and invasion capacity were markedly counteracted by LY294002, indicating that the PTEN/PI3K/AKT pathway rather than the PTEN/MAPK/ERK pathway was directly involved in the process of differentiation in SWO-38 cells (Fig. 6F).